基于EMD的多尺度形态学解调方法及其在机械故障诊断中的应用

为从机械故障信号中提取包含故障信息的特征频率，提出了基于EMD的多尺度形态学解调方法，该方法首先采用EMD方法将故障信号分解为有限个IMF分量，从中选取包含故障主要信息的IMF分量求和重构信号，再进行多尺度形态学解调，从而提取机械故障特征频率信息。将该方法用于滚动轴承、齿轮的故障诊断中，并与Hilbert包络方法比较，结果表明该方法能更好地提取故障特征频率，且对含噪故障信号也有较好的分析效果。

Application of multiscale morphology demodulation approach based on EMD in machinery fault diagnosis

A multiscale morphology demodulation approach based on EMD was proposed to extract the fault characteristic frequency from the mechanical failure signal. EMD method was used to decompose the failure signal into finite number of intrinsic mode functions(IMFs), and the signal was then reconstructed with a certain number of IMFs which contain the major failure informations. The multiscale morphology method was subsequently applied to the reconstructed signal, and the defect frequency informations were finally extracted. The method has been used in rolling bearing and gear drive fault diagosis, and the results were compared with those of the Hilbert envelop analysis. It is concluded that the proposed method is a better method for extracting the fault characteristic frequency, and also has a good effect on the noised failure signal analysis.